A Night at the Office #2

It's kinda difficult to effectively DevLog in retrospect... but there are a few things I wanted to note about the process that wouldn't necessarily come up if I started recording my progress from this point.

As mentioned in the first post, this whole exercise was inspired by the character my girlfriend created for a Vampire: The Masquerade campaign. You might be wondering "Why are you writing about your girlfriend's character, when you have one of your own?" The answer is simple: my character - Shane Ryan, the self-styled "Skeptical Paranormal Vlogger" - wasn't as immediately engaging in quite the same way.

Abigail Parker Winthrop was intriguing, and full of possibilities. Abigail is 40-something, but was turned at 30-something. She's able to pass off her youthful appearance as having 'good genes' for the moment - partly because it's true, and Grandmama Parker is a comparatively spritely woman in her 90s - but that won't work forever.

As a politician, of sorts, she'd be in the public eye almost by default, which would be difficult to manage for the traditional vampire, and particularly those from Vampire: The Masquerade, who are notoriously sun-averse. I don't know the 'canonical' story of her becoming a vampire - that was between my girlfriend (who doesn't remember) and our Storyteller, back in 2020 - so I ended up working hints of my own version of her backstory into the full version of the story. As a result, the sort of vampirism at play within the world of A Night at the Office is rather more sinister: it's the one where all of the legends about a vampire's weaknesses were created and propagated by the vampires themselves, to aid in their disguise. They can go out in sunlight, though prolonged exposure will harm them, and might eventually kill them; they have reflections, that sort of thing. They are more resistant to injury, but can be killed by normal means, not just a stake through the heart or decapitation. Their only serious weakness is that they can only feed on blood... and Abigail is a very picky eater.

Abigail is aware that Chicago's vampire overlords - a group of five or six, calling themselves the Potentates - consider her to be just another factotum, a pawn of sorts, mainly because her turning was bad luck (she had been in the wrong place at the wrong time, though there might be more to it than that) but she is useful to them in ways that their other peons could not even hope to be. While her condition precludes her from carrying on along her intended political trajectory, she has been installed as the head of an ostensibly healthcare-focussed caucus for Chicago's Republican party, the Red-Blooded Americans Caucus (R-BAC). Here, she is tasked with channeling the Potentates' interests into political policy.

Throughout all of this, she functions as a 'troubleshooter' for the Potentates, along with Shane and - while we were playing Vampire: The Masquerade - a couple of other vampires, called upon at a moment's notice to, just off the top of my head, escort a group of out-of-town vampires from the airport to the Prince's stronghold for some important negotiations, or investigate some gruesome murders in the vicinity of Lincoln Park Zoo. I currently have no plans to go into that area of her life, except tangentially, much later on.

A Night at the Office is just an introduction to Abigail and her world, and is deliberately 'small': a single location (mostly) with just two speaking characters (mostly), so that I can get to grips with the very basics of Ren'Py: variables, both session and persistent, ConditionSwitches, animated sprites, LayeredImages, timed choices, random choices, etc. It has ten potential endings, including one very bad end and one joke end, because I just couldn't resist. I already have ideas for two follow-up stories, which will be far less straightforward.

Of course, that relies on me finishing not only the demo of A Night at the Office, but the full story version... And while the demo is looking far more achievable now, the full story is still quite a way off. I've even got to the point where I've spent so long ignoring the main part of that story, that I can no longer remember the context of some of the features I'd added (some Achievements, in particular). What it does feature is one random choice feeding into a couple of ConditionSwitches, as well as a couple of Achievements relating to the single available ending.

As mentioned, Abigail will be facing a direct antagonist in the full story... but I don't want to give away exactly what that means. What I will say is that, while the 'Adult' content of the demo is fairly minimal and tame, it will be far more conspicuous in the full story. Access to the additional endings will be contingent on a long list of variables that will be set during what will essentially be a two-way interrogation... And that's the main sticking point for me at the moment.

A Night at the Office was originally conceived as minimally interactive, so I've been trying to rewrite it so that the information divulged in that dialogue now has to be acquired by asking the right questions... and the more you know about your antagonist, the more options you have for dealing with them. There is perhaps nothing more frustrating for a writer that 'fixing' the middle of the story. I've already tried several means of tracking the flow of the conversation (including - but not limited to - a flowchart and the Twine app) but I keep getting tangled up on how A leads to B and/or C, but D only comes out after B and C are exhausted, while E is only revealed by a specific question under C after D has been discussed. In the original draft of the story, A, B, C, D and E all came out because the antagonist kept talking.

Having figured out the Abigail sprite as a LayeredImage, the antagonist should be easy enough. They will have not only facial expressions but some animation as well. Nothing I haven't already done with Abigail, really, but on a grander scale and with more animations keyed to background variables. Abigail's expressions are built out of Brow, Eye and Mouth attributes, with the latter currently having two versions of each attribute, picked via a variable. The eyes are also set partially via a variable, but then blinking and other aspects are animated attributes of her LayeredImage. On top of this, she has clothing options which may or may not get further split into underwear, main clothing, and outerwear, depending on whether or not that seems necessary once I start patching Abigail into the full story. The change of artist, earlier this year, turned out fairly simple: as long as I got the filenames of the expression components right, they could slot right in. The only problems I had were with new components, or where I changed the way the clothing was handled.

I'll likely have the antagonist's sprite coding sorted before the story is ironed out, but I've no idea yet when I'll be in a position to commission that artwork. The two minor characters appearing in the demo are the priority, then their respective ending illustrations, which can then be reused in the full story. Most of the endings for the full story should be possible with just sprite animations, which I'm hoping to test once the minor character sprites are done.

There are already some persistent features in play, and the full story will have a couple of MultiPersistent variables that will affect the follow-up visual novel. The persistent features were kind of painful to implement: I had everything working perfectly in Ren'Py 7.5.3 (or so I thought), but started getting all kinds of error messages once I updated to Ren'Py 8.3.

It turned out that my main issue dated back to the 2020 hard disk crash, and the fact that I'd restarted development on a previously compiled version of my code. Once I started a new project in 8.3 and patched in the .rpy files, things started working properly... though I did later encounter a few issues with persistent data affecting the content of the title screen. Serves me right for trying to be clever, I guess.

Since I couldn't figure out what was happening, I ended up going back to the first functioning version of the demo - which I had backed up to the cloud, because I learn from my mistakes - and rewriting the parts that had gone wrong. I can't quite figure out where I went wrong, but I'm certainly glad to have it working again.